[georss] Multiple GeoRSS representations in one feed?

Allan Doyle adoyle at eogeo.org
Fri Oct 20 11:16:59 EDT 2006


On Oct 20, 2006, at 11:02, Carl Reed OGC Account wrote:

> An interesting conundrum. Perhaps what we need is the simple  
> ability for a
> RSS/GeoRSS payload to provide a URI or URL reference to an external  
> package
> of content structured as a GML document. I am having similar  
> discussions in
> the IETF and OASIS for the case in which a payload, such as CAP  
> (OASIS) or
> LoST (IETF) requires a very compact relatively simple expression of  
> location
> but then also has the requirement to reference much more complex  
> geometry,
> such as a floorplan. In both cases, neither standards organization  
> wants the
> complex (and large) GML document to be embedded in the initial  
> message. But
> they do want the ability for an application to know that there is a  
> larger,
> important document - so the recommended approach is to access that  
> document
> by reference
>
> Also, there are now many applications and product that can directly  
> ingest
> GML documents. Given that the GML serialization of GeoRSS is a GML
> application schema, there may already be any number of mapping/GIS
> applications that can already (or with minimal development) ingest  
> a GML
> document that is embedded in some other payload structure.

Funny. This is starting to sound a lot like the World Wide Web :)

In fact, the whole point of RSS is to be a summary annotation of  
something that is pointed to by the item.

RFC4287[1] has an example on page 4 showing the use of multiple  
<link> elements with different "rel" attributes. This is further  
explained in section 4.2.7:

4.2.7.  The "atom:link" Element

    The "atom:link" element defines a reference from an entry or feed to
    a Web resource.  This specification assigns no meaning to the  
content
    (if any) of this element.

    atomLink =
       element atom:link {
          atomCommonAttributes,
          attribute href { atomUri },
          attribute rel { atomNCName | atomUri }?,
          attribute type { atomMediaType }?,
          attribute hreflang { atomLanguageTag }?,
          attribute title { text }?,
          attribute length { text }?,
          undefinedContent
       }

I think for what might be considered "specialized" uses of GeoRSS,  
e.g. to have multiple representations and pointers to other things,  
it's appropriate to describe how this can be done. I'm not sure it's  
appropriate to toss in any more elements into GeoRSS itself to handle  
all the cases.

	Allan

[1] http://tools.ietf.org/html/rfc4287

>
> Cheers
>
> Carl
>
>
> ----- Original Message -----
> From: "Josh at oklieb" <josh at oklieb.net>
> To: "Pat Cappelaere" <pat at cappelaere.com>; <georss at lists.eogeo.org>
> Sent: Friday, October 20, 2006 7:33 AM
> Subject: Re: [georss] Multiple GeoRSS representations in one feed?
>
>
>> There is no particular restriction in Atom (or RSS) on including
>> multiple georss elements, or rather there isn't any way to provide
>> that restriction, except for recommending against the practice. It is
>> simply undefined what that communicates about the subject of the
>> entry / item. We could, I suppose, support a default "meaning" that
>> multiple georss elements in the same entry / item are to be
>> understood as exactly equivalent, alternate serializations.
>>
>> There is the possibility of expanding GeoRSS GML to define, e.g., the
>> scale at which to use a particular georss:where element, but I'm
>> pretty leery about such a complexity level without a very clear and
>> compelling need for it.
>>
>> Josh
>>
>> On Oct 20, 2006, at 9:24 AM, Pat Cappelaere wrote:
>>
>>> Andrew,
>>>
>>> I think you are right.
>>> I had raised the question at FOOS4G with Mikel:
>>> As a publisher, do I need to encode GeoRSS simple and Pro with the
>>> same
>>> entry? (actually, this could also include geo:lat/long to support
>>> Yahoo
>>> GeoRss)
>>>
>>> This is something that we ought to address ASAP.
>>> Pat.
>>>
>>>
>>>> From: Andrew Turner <georss at highearthorbit.com>
>>>> Date: Fri, 20 Oct 2006 09:09:27 -0400
>>>> To: Pat Cappelaere <pat at cappelaere.com>
>>>> Cc: Allan Doyle <adoyle at eogeo.org>, <georss at lists.eogeo.org>
>>>> Subject: Re: [georss] Multiple GeoRSS representations in one feed?
>>>>
>>>> I think Allan's question was about publishing multiple feeds using
>>>> different formats. So Jason, for example, could publish a Simple
>>>> using
>>>> points, and then a more complex using GML or Simple with lines.  
>>>> Both
>>>> feeds would have the same items, just different representations of
>>>> the
>>>> geographic data.
>>>>
>>>> What actually would be interesting, is if there is any problem
>>>> publishing 2 representations in the same feed? RSS parsers ignore
>>>> items they don't understand, so the Simple point readers would  
>>>> ignore
>>>> the GML. And perhaps a more featured GML parser should consume the
>>>> "highest fidelity" geometry from a single item. So the GML ring  
>>>> would
>>>> override the Simple point.
>>>>
>>>> Any particular problems with this? Or is it just confusing?
>>>>
>>>> The potential problem with multiple feeds w/ different
>>>> representations
>>>> is that if a user wants to switch the feeds from Simple to GML in
>>>> their reader, are the "read/unread" kept (GUID's the same), etc.
>>>>
>>>> Andrew
>>>>
>>>>
>>>> On 10/20/06, Pat Cappelaere <pat at cappelaere.com> wrote:
>>>>> Allan,
>>>>>
>>>>> It is not unusual for us to display more than one feed on the
>>>>> map.  They are
>>>>> displayed on different layers so they can be turned on/off.
>>>>> This is really no different than feature layers.
>>>>> Pat.
>>>>>
>>>>>
>>>>>> From: Allan Doyle <adoyle at eogeo.org>
>>>>>> Date: Fri, 20 Oct 2006 08:30:38 -0400
>>>>>> To: <georss at lists.eogeo.org>
>>>>>> Subject: Re: [georss] Multiple GeoRSS representations in one  
>>>>>> feed?
>>>>>>
>>>>>> I guess I have to wonder also to what extent it's bad to have two
>>>>>> feeds? If we try to cram everything into one representation,
>>>>>> then we
>>>>>> climb up the complexity wall again and leave the simple people
>>>>>> (like
>>>>>> me) behind.
>>>>>>
>>>>>> Allan
>>>>>>
>>>>>> On Oct 20, 2006, at 07:23, Raj Singh wrote:
>>>>>>
>>>>>>> As Ron said, holes in polygons aren't supported in simple,  
>>>>>>> but are
>>>>>>> possible in GeoRSS GML. You'll still have the chicken and egg
>>>>>>> issue
>>>>>>> that you're not likely to find many clients (that aren't "real"
>>>>>>> mapping systems) that will correctly render them. That, and
>>>>>>> aggregates, begin to stretch the limits of GeoRSS. If those
>>>>>>> sophisticated geometries are key to what you're doing, you may
>>>>>>> be in
>>>>>>> the market for a heavier standard, like GML Simple Features
>>>>>>> Profile,
>>>>>>> listed on this page <http://www.opengeospatial.org/standards/ 
>>>>>>> gml>.
>>>>>>> ---
>>>>>>> Raj
>>>>>>>
>>>>>>>
>>>>>>> On Oct 20, 2006, at 8:52 AM, Jason Birch wrote:
>>>>>>>
>>>>>>>> Is there any way to represent donut polygons in simple?  I can
>>>>>>>> just
>>>>>>>> provide the exterior ring, but it would be nice to put holes
>>>>>>>> where
>>>>>>>> they belong.
>>>>>>>>
>>>>>>>> And aggregates... I have quite a few parcels that were split by
>>>>>>>> roads, and are actually part of the same legal feature.  I  
>>>>>>>> don't
>>>>>>>> really need to have them joined as a MultiGeometry, but I can't
>>>>>>>> tell from the XSD whether multiple geometries are allowed  
>>>>>>>> inside
>>>>>>>> the "where".
>>>>>>>>
>>>>>>>> Jason
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> georss mailing list
>>>>>>> georss at lists.eogeo.org
>>>>>>> http://lists.eogeo.org/mailman/listinfo/georss
>>>>>>
>>>>>> --
>>>>>> Allan Doyle
>>>>>> +1.781.433.2695
>>>>>> adoyle at eogeo.org
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> georss mailing list
>>>>>> georss at lists.eogeo.org
>>>>>> http://lists.eogeo.org/mailman/listinfo/georss
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> georss mailing list
>>>>> georss at lists.eogeo.org
>>>>> http://lists.eogeo.org/mailman/listinfo/georss
>>>>>
>>>>
>>>>
>>>> -- 
>>>> Andrew Turner
>>>> ajturner at highearthorbit.com        42.4266N x 83.4931W
>>>> http://highearthorbit.com              Northville, Michigan, USA
>>>
>>>
>>> _______________________________________________
>>> georss mailing list
>>> georss at lists.eogeo.org
>>> http://lists.eogeo.org/mailman/listinfo/georss
>>
>> _______________________________________________
>> georss mailing list
>> georss at lists.eogeo.org
>> http://lists.eogeo.org/mailman/listinfo/georss
>
> _______________________________________________
> georss mailing list
> georss at lists.eogeo.org
> http://lists.eogeo.org/mailman/listinfo/georss

-- 
Allan Doyle
+1.781.433.2695
adoyle at eogeo.org






More information about the georss mailing list